Large scale genome-wide association studies, for the first time, have identified genetic variation definitively associated with major depression. To translate this advancement into improved diagnosis, monitoring, and treatment, a critical next step is to elucidate the behavioral mechanisms linking the implicated genetic variation with depression. Unfortunately, the large-scale studies that have identified associated variants have typically employed single-time point and limited phenotypic assessments that are not suited to study mechanisms linking genes and depression, a chronic multi-modal disease. Our long-term goal is to elucidate the pathophysiological architecture underlying depression to facilitate the development of improved treatments. Our objective in this application is to understand how genetic variants associated with the development of depression exert their effect. Medical internship, the first year of professional physician training, presents a unique situation in which we can prospectively predict the onset of a uniform, chronic stressor and follow the development of depressive symptoms. We have found that rates of depression increase dramatically, from 4% prior to internship to 26% during internship year. Currently, the study enrolls 3,000-3,500 interns annually. Our intern cohort is an ideal population to closely monitor the development of depression with recent mobile health technology as a tool to follow these individuals in real-time, with objective measures. In the proposed study, we will combine, cutting edge-genomics, mobile health technology, and the prospective intern stress design to identify the mechanisms through which depression-related genetic variation lead to depression. We hypothesize that depression-associated genetic variation acts to increase the risk of depression through specific mobile measured behavioral phenotypes. To test this hypothesis, we propose the following three specific aims: 1) Identify data driven behavioral phenotypes, derived from mobile data elements, that predict short-term risk for mood changes and depressive episodes; 2) Identify genetic variants associated with depression under stress; and 3) Elucidate behavioral phenotypes through which genetic variants may act to increase the risk of depression. Our approach is innovative because it combines a naturally occurring stress paradigm and new real-time objective assessment tools in order to elucidate the relationship between genes, objective, real-time markers and depression with an approach that, to date, has not been attempted. This project is significant because it has the potential to identify key mechanisms underlying genetic associations involved in depression under stress, an advancement that holds promises in predicting treatment response and identifying novel targets for antidepressant development.